September 2018 / No. 56 1

Observations on market abuse surveillanceIn Market Watch issues 48 and 50, we discussed our observations from a series of suspicious transaction reporting (STR) supervisory visits. These included the calibration of firms’ surveillance systems, where we had seen some cases of firms’ over-reliance on ‘out of the box’ alert calibration. Another topic we covered was the use of market abuse risk assessments, particularly the benefits of firms undertaking a detailed assessment of the risks to which firms are exposed before designing a surveillance programme. We also explained that surveillance appeared to be less developed for some asset classes, meaning it was possible that instances of potential market abuse were not being identified. Since the publication of Market Watch 48 in 2015 and Market Watch 50 in 2016, we have continued to visit firms and trading venues to assess their market abuse surveillance arrangements. We consider it is now helpful to revisit these topics, as well as discussing some fresh observations.

Calibration of surveillance systemsWe have continued to observe firms with vendor-supplied systems using ‘out of the box’ and ‘industry standard’ settings to calibrate their alert parameters. We have also observed firms using average peer alert volumes as a measure of the appropriateness of their calibration. While we recognise that firms may see certain benefits in understanding how their approach relates to that of their peers, putting undue weight on these comparisons creates risks to the independent assessment of each firm’s business.

We remind firms that every business (and every firm’s client base) will be unique, and therefore different to its peers’ in some respects. Therefore, relying on peer standards, such as popular ‘out of the

ContentsObservations on market P1 abuse surveillance

Payment for Order Flow (’PFOF’) P3

If you wish to join our email list to receive future editions, please contact us at market.watch@fca.org.uk

You can also find issues on our website at: www.fca.org.uk/firms/markets/market-abuse

FCA regulated fees and leviesMarket WatchNewsletter on market conduct and transaction reporting issues

September 2018

Market Watch

September 2018 / No. 56 2

box’ alert settings, average peer parameters and average peer output volumes, will not necessarily satisfy MAR requirements. In particular, firms may not meet the requirement that each firm’s surveillance arrangements, systems and procedures are appropriate and proportionate to the scale, size and nature of their business activity. Each firm is responsible for making its own judgements about alert calibration, and firms risk failing to comply with MAR if they assume that because a certain calibration is appropriate for their peers, it must be appropriate for them.  

Assessing the risk of market abuseWe have observed firms referring in market abuse risk assessments to the list of indicators for fictitious devices, false or misleading signals and price securing in MAR (and the list of related practices in the level two legislation1) and treating those lists as exhaustive. We remind firms that the lists in MAR are not exhaustive; firms treating them as such may fail to identify the risk of, and so fail to detect and report, other types of market manipulation which are still within the broader scope of MAR article 12(1)(a) and (b). Similarly, we remind trading venue operators of the requirement to consider signals not specifically listed in the MiFID2 level two legislation2 when they design surveillance to detect and report possible market abuse.  

We have observed how some firms consider potential risks for each business area and from end-to-end of the whole product or service cycle. We note that risks identified in one area through this bottom-up exercise often arise elsewhere, which underlines the benefit of sharing analysis across the business. All risks can then be aggregated in risk frameworks or risk and control self-assessment models.

Fixed income surveillance STOR submission across asset classes remains inconsistent and we believe submissions are lower than they should be in some areas. In particular, our view is that submissions continue to be too low in fixed income products and we wish to provide some further observations from our recent visit programme where we have focused on fixed income markets.

In Market Watch 50, we explained that firms are often over-prescriptive with analysts and do not encourage them to look beyond the initial alert. Our observations indicate this continues to be the case with some firms. In some fixed income markets, for example, some analysts tended to take a narrow approach, reviewing only the activity in the product which triggered the alert and not considering other trading in correlated products. Because many fixed income products are inter-connected, consideration of trading activity in correlated products - such as cash vs futures, or products with different durations - is an important element of effective surveillance.

We have observed firms using price-driven surveillance for products where yield is the primary basis on which pricing and trading is undertaken. If firms do not use yield in either the derivation of alerts for these products or in their review, they may fail to carry out meaningful monitoring. We remind firms that effective market abuse surveillance requires that the techniques and methodologies used are appropriate to the products under surveillance.

1 Commission Delegated Regulation (EU) 2016/522, Annex II2 Commission Delegated Regulation (EU) 2017/565, Annex III Section B

Market Watch

September 2018 / No. 56 3

We have observed some firms facing challenges in accessing transaction data for less liquid fixed income instruments, because of the relative infrequency of trades in these markets. This can potentially result in analysts being unable to undertake adequate surveillance of these instruments. Where these challenges remain, we have observed examples of firms using effective tactical solutions, such as the use of tools and metrics that may be more readily available in the absence of transaction data. This included further analysis of the following: large trades; trades that resulted in large positions in an instrument; trades that were not booked in a timely manner; trades in related instruments; orders and trades in instruments with pre-existing large positions.

Firm rationales for failingsWe have observed several firms using questionable rationales to reconcile themselves to their potential failure to meet their obligations under MAR. Two themes are particularly common here.

Firstly, some firms appear to consider that their own failings can be excused by a perception that some of their peers are failing in the same way. While we have on occasion acknowledged that industry in general faces specific challenges, we will not necessarily accept failures to comply with MAR on the basis that multiple firms are in a similar position. Equally, where we have not yet taken public enforcement action against a particular failing at one firm, other firms should not assume we will not take action against them for similar failings. We decide how we resolve failures to comply with MAR based on a number of public and non-public factors. Many of these factors will vary between firms, even apparently similar firms within the same industry peer group.

Secondly, on STOR visits where we observe potential failings, firms occasionally tell us that the responsible employee, for example the firm’s CF10/SM16, has only recently joined and does not feel they are responsible for a predecessor’s arrangements. We remind firms that on STOR supervisory visits we are primarily reviewing the firm’s compliance with the STOR regime, rather than employees’ compliance with their individual regulatory obligations. Accordingly, on STOR supervisory visits, we consider the time an employee has been in their role to be of limited relevance, and we will not generally treat it as mitigating the firm’s failings.

Payment for Order Flow (’PFOF’)

Introduction Payment for Order Flow (PFOF) occurs when an investment firm (typically a broker) that executes orders on behalf of its client receives a fee/commission from the client that originates the order, as well as from the counterparty the trade is then executed with (typically a market maker or other liquidity provider). These payments create a conflict of interest between the firm and its client by incentivising the firm to execute its client orders with counterparties willing to pay the highest commission and so undermine the firm’s ability to act as a good agent. This practice restricts transparency and efficiency in the price formation process. It also distorts competition

Market Watch

September 2018 / No. 56 4

by forcing liquidity providers to use a pay-to-play model.

In December 2017, the FCA published a Dear CEO letter, reconfirming our position on PFOF. The letter also set out our expectations on the practice in the context of MiFID II, particularly its strengthened conflicts of interest regime. Subsequently, our Business Plan 2018/2019 set out supervisory work on PFOF as one of our priorities to address conflicts of interest in the wholesale sector. We said this work would focus on ensuring that firms are complying with the strengthened standards in MiFID II. We also said we would assess whether the rules on conflicts of interest and best execution are working as intended.

This article provides an update on our most recent work on PFOF, including:

• sharing high-level findings of our supervisory work to date and

• setting out the further work and our planned next steps in this area.

Focus of our workBefore the implementation of MiFID II, we found that most firms had stopped charging PFOF for retail and professional client business. However, market intelligence suggested that some firms were still charging PFOF for ECP business, notwithstanding our concern that this practice is inconsistent with their obligations to manage conflicts of interest. Intelligence also suggested that some brokers were considering various avoidance tactics so they could continue to charge PFOF after MiFID II implementation. This was the context for our ‘Dear CEO’ letter on PFOF in December 2017.

As a result, our current supervisory work has focused on PFOF charged by firms for ECP business, following the implementation of MiFID II.

Findings to dateOur supervisory work to date suggests that there have been several developments in the market since January 2018. Some of our early findings from visits include:

• Nearly all of the brokers we have visited have now stopped charging PFOF where they consider themselves as acting in an agency-like capacity, regardless of the client’s categorisation. Feedback from market makers confirms this finding. As and when we identify outliers, we will require them to stop and consider retrospective action.

• Many of the brokers we have visited still charge both sides of a transaction for what they characterise as interdealer-broking business, where they do not consider themselves to be acting in an agency capacity. These instances are now the vast majority of transactions where brokers charge both sides of a transaction. We are currently gathering more information about brokers’ activities for this type of business. This will enable us to assess whether firms are applying consistent judgements across the market that align with the rules on managing conflicts of interest.

Market Watch

September 2018 / No. 56 5

• We have found isolated evidence of behaviour, in the context of providing an agency-like service to clients, whereby a broker seeking liquidity from liquidity providers has been wrongly defined as a service to those counterparties, with a corresponding charge being levied on the liquidity provider. We will take action against firms who characterise their relationship with liquidity providers in a way that does not reflect economic reality and which are PFOF arrangements.

• We have received reports of some brokers booking transactions to overseas offices so they could argue that they were allowed to charge PFOF. We will continue to examine brokers’ order routing to overseas entities for any evidence of circumventing behaviour that breaches our rules; for example, when relevant orders are handled at any stage by a UK broker but are subject to two-sided charging contrary to our PFOF position.

• We have not found evidence that brokers have implemented schemes or designed structures to circumvent the rules. Examples of behaviour that would indicate this include linking charges for market makers to non-execution services, such as research products, market analysis software or requiring alternative order flows to replicate PFOF.

Next stepsWe will continue our programme of firm visits. These will assess how robust firms’ systems and controls are for monitoring adherence to all relevant rules, policies and procedures on PFOF. Firms should note that, for a given transaction, they should analyse the capacity in which they and the different counterparties involved act, rather than look to their or these counterparties’ general business models. Therefore our work will include scrutiny of the specific controls for correctly classifying individual transactions, such as what a firm classifies as interdealer broking activities. We will also look at the policies applied by firms, focusing particularly on how they manage potential conflicts, and the extent of their compliance monitoring and oversight activity.

Once we have analysed our findings, we will consider further action as appropriate.

We will also continue to engage with our European counterparts on the issue of PFOF to ensure a level playing-field for firms across the EEA.